This invention relates to redundant storage of data in the form of logical volumes in a plurality of automated data storage libraries, and, more particularly, to the recall of logical volumes to cache from physical media volumes in a library, where the logical volumes are to be copied into another library for redundant storage.
Data processing systems typically require large amounts of data storage capacity, some of which is needed quickly and may be stored in memory and hard disk drives, and other of which is not immediately required. As an example, data not immediately required may comprise data that is infrequently accessed, and the storage of the data may be in the form of logical volumes of data stored on removable rewritable physical media volumes, such as magnetic tape or optical disk, and the physical media volumes may be written and/or read by means of a data storage drive.
If large amounts of data are to be stored and then accessed on occasion, automated data storage libraries are often employed. Such libraries provide efficient access to large quantities of data stored on physical media volumes which are stored in storage shelves and which are accessed by one or more accessors and delivered to data storage drives in the library.
It is desirable to provide a level of redundancy of the stored data to provide constant availability of access to the data, even should an automated data storage library or a communication path to an automated data storage library become unavailable. An example of a data storage library system for redundantly storing and accessing data stored as logical volumes on physical media volumes in a plurality of automated data storage libraries is the IBM Magstar Peer-to-Peer Virtual Tape Server, which comprises a plurality of directors, such as xe2x80x9cAX0xe2x80x9d Virtual Tape Controllers, and at least two automated data storage libraries, such as 3494 Magstar Virtual Tape Servers.
A request by a host data processing system to create or modify a logical volume is steered by a director to one of the libraries. If the request requires access to a physical media volume that contains the requested logical volume, the library accesses the physical media volume from its storage shelf, and mounts the physical media volume at a desired data storage drive. The logical volume is read to cache storage, which may comprise hard disk drives or other high speed storage, so that it may be immediately accessed and provided to the host system. If the request is for a logical volume that is already in cache, or is for a logical volume that will be completely rewritten (scratch mount), a physical media volume access is not required. The host system then reads from or writes to the logical volume in the cache, via a director. When the host system closes the logical volume, the logical volume remains in cache storage so that it can be immediately reaccessed. If the logical volume was updated, the directors will typically immediately re-access the logical volume from the cache storage of one library and copy it to the other library so that both libraries have the most current copy.
The cache storage is typically limited in capacity, requiring that the updated logical volumes be migrated to backing storage, comprising the physical media volumes, so as to free space in the cache storage. Typically, a least recently used (LRU) algorithm is employed to migrate logical volumes out of cache storage to backing storage. Preferably, migration of a logical volume occurs subsequent to the re-access and copying of the logical volume to the other library.
Should one of the libraries become temporarily unavailable or the communication path from the directors to the library become temporarily unavailable, for example, because of a service action, the remaining library will continue to provide logical volumes and to accumulate updated logical volumes, without an opportunity to immediately copy an updated logical volume to the other library. Thus, the copying is deferred to a point when the other library again becomes available. Alternatively, a customer may select a deferred copy mode in order to provide higher peak host system input/output performance.
Should the deferred copy mode continue for an extended period of time, for example, several hours, many of the logical volumes which are to be copied must be migrated to backing storage so as to free space in the cache storage.
Recalling and copying a migrated logical volume requires that the physical media volume containing the migrated logical volume be mounted on a data storage drive, so that the logical volume may be recalled into cache storage, re-accessed and copied. Mounting a physical media volume is so time consuming that it may be difficult, or impossible, to catch up the copying of all logical volumes whose copying has been deferred.
An object-of the present invention is to increase the efficiency of the recall and copying of migrated logical volumes.
An automated data storage library, multiple library system, method, and computer program product are provided for recalling logical volumes of data stored by an automated data storage library in response to a request to copy a logical volume that has been migrated, the request received, for example, from one of a plurality of directors. Each automated data storage library stores logical volumes, including logical volumes to be copied to a second automated data storage library, and the copy request for a migrated volume requires recalling the logical volume to be copied. Hence, herein, the terms, xe2x80x9ccopy requestxe2x80x9d, xe2x80x9crecallxe2x80x9d, xe2x80x9crecall requestxe2x80x9d, etc., are used interchangeably.
The library comprises a plurality of storage shelves for storing physical media volumes, wherein the physical media volumes are each capable of storing a stacked plurality of the logical volumes. At least one data storage drive is provided for reading and/or writing the stacked logical volumes of the physical media volumes, and at least one accessor is provided for transporting the physical media volumes amongst the storage shelves and the data storage drives and for mounting and demounting the physical media volumes at the data storage drives. A cache storage is provided for storing a subset of the logical volumes, and a library controller is provided for operating the library in accordance with the present invention, including the migration of logical volumes from the cache storage to the physical media volumes as stacked logical volumes, to free up space in cache storage.
In accordance with the present invention, the library controller responds to a recall of a logical volume to be copied, which logical volume has been-migrated to the physical media volumes, by first determining which physical media volume contains the requested migrated logical volume. The library controller also identifies all of the logical volumes that need to be copied, in addition to the requested migrated logical volume, that are migrated and stored as stacked logical volumes on the determined physical media volume containing the requested migrated logical volume. The library controller then selects up to xe2x80x9cNxe2x80x9d of the identified logical volumes, including the requested migrated logical volume, and arranges a selection list of the selected logical volumes optionally, the list of selected logical volumes is arranged in a sequence in accordance with the order that the selected logical volumes are written on the determined physical media volume. The library controller employs the accessor to transport and mount the determined physical media volume to a data storage drive, recalling the selected logical volumes of the mounted physical media volume, via the data storage drive, to the cache storage in accordance with the selection list. Then, in response to the recall request, the library controller provides a list of the recalled selected logical volumes to one, or a plurality of, directors.
The library controller optionally gives priority to the identified logical volumes most closely adjacent the requested migrated logical volume in selecting the up to xe2x80x9cNxe2x80x9d identified logical volumes. In this manner, the logical volumes best comprising a closely positioned sequence are recalled together at one time, saving substantial accessing time.
For a fuller understanding of the present invention, reference should be made to the following detailed description taken in conjunction with the accompanying drawings.